Computational and Systems Biology

Role of desolvation on biomolecular liquid-liquid phase separation

Kai Zhang
Zhiyu Peng
Wenfei Li author has email address
Wei Wang author has email address

Department of Physics, National Laboratory of Solid State Microstructure, Nanjing University, Nanjing, China
Wenzhou Key Laboratory of Biophysics, Wenzhou Institute, University of Chinese Academy of Sciences, Wenzhou, China
Jiangsu Key Laboratory for Cardiovascular Information and Health Engineering Medicine, Department of Cardiology, Nanjing Drum Tower Hospital, Medical School, Nanjing University, Nanjing, China

https://doi.org/10.7554/eLife.111124.1

Open access
Copyright information

Figures and data

Desolvation effect of inter-molecular interactions.
(A) Potential of mean force along the inter-molecule distances for the amino acid analogues from all-atom MD simulations with different water polarity. (B) Schematic diagram of desolvation effect. (C) Pair-wise energy function incorporating desolvation terms. Different curves correspond to different desolvation parameters.

Thermodynamic regulation and microscopic mechanisms of desolvation-mediated phase separation.
(A) Baseline phase diagram of the poly-50 system using the standard HPS model. (B) Representative simulation snapshots visualizing the transition from a stable condensate (T ^∗ = 2.58) to a near-critical state (T ^∗ = 2.98) and a homogeneous solution (T ^∗ = 3.18). (C) Time-averaged density profiles along the z-axis identifying the coexisting dense and dilute phases. (D, E) Macroscopic phase boundaries under varying desolvation barrier heights ϵ_b (D) and solvent-separated potential depths ϵ_ss (E). Insets show the monotonic dependence of T_c on the respective parameters. The lower schematics highlight the distinct thermodynamic drivers: entropic penalty (ϵ_b) versus enthalpic stabilization (ϵ_ss). (F, G) Renormalized phase behavior plotted against normalized temperature T ^∗/T_c for varying ϵ_b (F) and varying ϵ_ss (G).

Effect of desolvation on protein conformations.
(A) Schematic illustration of the conformational distributions of the protein in the high-and low-density phases under different desolvation parameters. (B-C) Distribution of R_g with different ϵ_b (B) and ϵ_ss (C) at . The solid and dashed lines represents the results in the condensed and dilute phases respectively. The inset illustrates the mean value of R_g as a function of desolvation parameters. (D) Correlation between temperature difference and the averaged R_g difference between two phases. The purple and yellow dashed lines represent linear fits to the data obtained at low simulation temperatures and high-temperature regimes respectively. The inset displays the improved linearity obtained when the temperature difference is normalized by the simulation temperature, . (E) Correlation between the density difference and R_g difference in the two phases with varying temperatures and desolvation parameters. The plot was shown in a log-log scale.

Effect of desolvation on protein conformations.
(A) Schematic illustration of the conformational distributions of the protein in the high-and low-density phases under different desolvation parameters. (B-C) Distribution of R_g with different ϵ_b (B) and ϵ_ss (C) at . The solid and dashed lines represents the results in the condensed and dilute phases respectively. The inset illustrates the mean value of R_g as a function of desolvation parameters. (D) Correlation between temperature difference and the averaged R_g difference between two phases. The purple and yellow dashed lines represent linear fits to the data obtained at low simulation temperatures and high-temperature regimes respectively. The inset displays the improved linearity obtained when the temperature difference is normalized by the simulation temperature, . (E) Correlation between the density difference and R_g difference in the two phases with varying temperatures and desolvation parameters. The plot was shown in a log-log scale.

Desolvation-mediated modulation of diffusion and coarsening dynamics.
(A) Snapshots of spinodal decomposition and equilibrium chain diffusion. The top panels illustrate domain coarsening, while the bottom panels highlight self-diffusion of highlighted chains in the stable slab. The inset quantifies the chain mobility via TAMSD analysis. (B) Density profiles of biomolecules along the z-axis for the entire system (orange) and for the highlighted chains (red) at different time lags with and without desolvation. The solid curves represent normalized averages over 1 μs. (C) Diffusion coefficients as a function of the inter-chain desolvation strength (ϵ_b, ϵ_ss) and dense-phase densities (ρ_dense) at different temperatures. (D) Reduced diffusion coefficient versus quench depth for varying ϵ_b. The schematic inserts highlight the impact of energy landscape roughness on diffusion dynamics. (E) Schematic representation of the phase separation mechanism. The diagram depicts the transition from initial density fluctuations to late-stage growth, mediated by an intermediate plateau. The zoom-in view details the structural origin of the kinetic arrest arising from strong inter-chain interactions. (F, G) Kinetics of density fluctuations and domain growth. Top: Time evolution of the normalized density variance The inset shows the characteristic time τ_1/2. Bottom: Growth of the average slab size ⟨ξ⟩ on a logarithmic scale. The timeline bar highlights the three distinct regimes: initial fluctuation, kinetic arrest (plateau), and late-stage coarsening via Brownian Motion Coalescence (BMC).

Desolvation-mediated modulation of diffusion and coarsening dynamics.
(A) Snapshots of spinodal decomposition and equilibrium chain diffusion. The top panels illustrate domain coarsening, while the bottom panels highlight self-diffusion of highlighted chains in the stable slab. The inset quantifies the chain mobility via TAMSD analysis. (B) Density profiles of biomolecules along the z-axis for the entire system (orange) and for the highlighted chains (red) at different time lags with and without desolvation. The solid curves represent normalized averages over 1 μs. (C) Diffusion coefficients as a function of the inter-chain desolvation strength (ϵ_b, ϵ_ss) and dense-phase densities (ρ_dense) at different temperatures. (D) Reduced diffusion coefficient versus quench depth for varying ϵ_b. The schematic inserts highlight the impact of energy landscape roughness on diffusion dynamics. (E) Schematic representation of the phase separation mechanism. The diagram depicts the transition from initial density fluctuations to late-stage growth, mediated by an intermediate plateau. The zoom-in view details the structural origin of the kinetic arrest arising from strong inter-chain interactions. (F, G) Kinetics of density fluctuations and domain growth. Top: Time evolution of the normalized density variance The inset shows the characteristic time τ_1/2. Bottom: Growth of the average slab size ⟨ξ⟩ on a logarithmic scale. The timeline bar highlights the three distinct regimes: initial fluctuation, kinetic arrest (plateau), and late-stage coarsening via Brownian Motion Coalescence (BMC).

Parameterization of desolvation terms for the HPS model and CALVADOS2 model based on IDPs.
(A) Schematic workflow of the desolvation parameterization. (B) Correlation between experimental R_g and simulation R_g for the original HPS model (blue) and the revised HPS model with default desolvation scales (α = 0.33 and α = 0.06) (purple). The for different models were also shown. (C) Correlation between experimental R_g and simulation R_g for the original CALVADOS2 model (orange), the revised CALVADOS2 model with default desolvation (yellow), and revised CALVADOS2 model with optimized desolvation (α_b = 0.3, α_ss = 0.03 and ϵ = 0.262 kcal/mol)(red). (D) Coexistence curves of FUS LC simulated with the original HPS model (blue), the revised HPS model with default desolvation scales (purple), the energy-rescaled HPS model (green), the default CALVADOS2 model (orange), and the revised CALVADOS2 model with optimized desolvation scales (red). The green shaded regions highlight the deviations of the desolvation models relative to the original frameworks.

Parameterization of desolvation terms for the HPS model and CALVADOS2 model based on IDPs.
(A) Schematic workflow of the desolvation parameterization. (B) Correlation between experimental R_g and simulation R_g for the original HPS model (blue) and the revised HPS model with default desolvation scales (α = 0.33 and α = 0.06) (purple). The for different models were also shown. (C) Correlation between experimental R_g and simulation R_g for the original CALVADOS2 model (orange), the revised CALVADOS2 model with default desolvation (yellow), and revised CALVADOS2 model with optimized desolvation (α_b = 0.3, α_ss = 0.03 and ϵ = 0.262 kcal/mol)(red). (D) Coexistence curves of FUS LC simulated with the original HPS model (blue), the revised HPS model with default desolvation scales (purple), the energy-rescaled HPS model (green), the default CALVADOS2 model (orange), and the revised CALVADOS2 model with optimized desolvation scales (red). The green shaded regions highlight the deviations of the desolvation models relative to the original frameworks.

Desolvation parameters α_b and α_ss derived by fitting all-atom MD simulations.
The optimized parameter ϵ for CALVADOS2 is also listed together with that used in the HPS model.

Summary of the production run parameters for all-atom molecular dynamics simulations.

Summary of the coarse-grained molecular dynamics simulation parameters used in the slab simulations.

Desolvation effect from all-atom MD simulations.
(A-D) PMFs from all-atom simulations and the fitting with desolvation energy function (Eq. 1) for different pairs of amino acid analogues at physiological salt concentration. (E) The collection of the fitted PMFs for different pairs of amino acid analogues. (F) The extracted α_b and α_ss values for different pairs of amino acid analogues.

Effect of desolvation on the radial distribution of the residue pairs of the homopolymer chains.
(A, B) Schematic figures for the change of desolvation potential. (C, D) The radial distribution functions between inter-chain and intra-chain residue pairs for simulations with different ϵ_b and different ϵ_ss at the same temperature (T ^∗ = 1.49). (E, F) The radial distribution functions between all residue pairs in the dense phase for simulations with different ϵ_b and different ϵ_ss at the same temperature ()

Effect of desolvation on the radial distribution of the residue pairs of the homopolymer chains.
(A, B) Schematic figures for the change of desolvation potential. (C, D) The radial distribution functions between inter-chain and intra-chain residue pairs for simulations with different ϵ_b and different ϵ_ss at the same temperature (T ^∗ = 1.49). (E, F) The radial distribution functions between all residue pairs in the dense phase for simulations with different ϵ_b and different ϵ_ss at the same temperature ()

Optimizing the desolvation parameters for the CALVADOS2 model by using experimental R_g data.
(A,C,E,G) Correlation between experimental R_g values and simulation R_g values with different desolvation parameters. (B,D,F,H) values as a function of for different sets of desolvation parameters. The optimal ϵ values are also shown.

Optimizing the desolvation parameters for the CALVADOS2 model by using experimental R_g data.
(A,C,E,G) Correlation between experimental R_g values and simulation R_g values with different desolvation parameters. (B,D,F,H) values as a function of for different sets of desolvation parameters. The optimal ϵ values are also shown.

Correlation between experimental R_g values and simulation R_g values with different α_b and α_ss values and the corresponding optimized ϵ values for the CALVADOS2 model.

Comparison of single chain radius of gyration (R_g) for HPS model, HPS + desolvation, HPS (ϵ rescaled to 0.16 kcal/mol), CALVADOS2 model and CALVADOS + desolvation.

Sign up for email alerts